|Publication number||US4193096 A|
|Application number||US 05/784,206|
|Publication date||Mar 11, 1980|
|Filing date||Apr 4, 1977|
|Priority date||Apr 4, 1977|
|Publication number||05784206, 784206, US 4193096 A, US 4193096A, US-A-4193096, US4193096 A, US4193096A|
|Inventors||James C. Stoffel|
|Original Assignee||Xerox Corporation|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (9), Referenced by (40), Classifications (7)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This invention relates to a system for compressing or compacting image data, and more particularly, to a system for compressing image data utilizing predictive encoding.
Pictorial information may be converted to binary data for various operating purposes as for example, to enhance transmittal of the pictorial information from one location to another, for storage purposes etc. If the data can also be compressed or compacted, certain highly desirable advantages accrue. For example, the amount of time required for transmittal of the data from one location to another can be greatly reduced. Additionally, existing data storage facilities can also be used more efficiently.
While therefore substantial benefits arise from data compaction, high data compaction ratios are of no avail if the pictorial information represented by the compressed data, is distorted or inaccurate. Optimumly, the highest compaction ratio without distortion of the original data is required.
It is therefore a principal object of the present invention to provide a new and improved system for pictorial data compression.
It is an object of the present invention to provide a system for accurately compacting pictorial data utilizing improved predictive encoding techniques.
It is an object of the present invention to provide method and apparatus for reducing storage/bandwidth required to store pictorial data.
It is an object of the present invention to provide a relatively simple and improved all digital system for accurately encoding a picture using a reduced number of data pixels.
It is an object of the present invention to provide a new and improved process for converting pictorial data into coded form for transmittal and storage having reduced bandwidth.
This invention relates to a system for compressing pixel data representative of an image, comprising in combination, means for scanning an original document to produce an image pixel pattern representative of the image on the document; means for screening the image pixel pattern to produce binary pixel data representative of the image pixel pattern; means for arbitrarily subdividing the binary pixel data into a succession of discrete areas, each of the discrete areas being of predetermined size; means for establishing an image value for the discrete areas; predictor means for predicting the image value of each discrete area from the established image value of at least one adjacent discrete area; comparator means for comparing the predicted image value of each discrete area with the established image value for the discrete area; error code generating means for generating an error code representative of the degree of error found by the comparator means; and encoding means for encoding the output of the error code generating means to produce a compressed representation of the image.
This invention further relates to a method of compressing a stream of digital data representative of an image, the steps comprising, scanning an original document to produce a pixel pattern representative of the image on the document; screening the image pixel pattern to convert the image pixel pattern from an analog pixel pattern to a binary pixel pattern; subdividing the binary pixel pattern into a succession of discrete areas each containing a preset number of digital bits; establishing an image value for the discrete areas; predicting the image value for the discrete areas from the established image value of at least one adjoining area; comparing the predicted image value of each discrete area with the established image value of that area and generating an error signal representative of the size of the error between predicted and established image values; and encoding the error signals to provide a stream of digital signals representative of the image.
Other objects and advantages will be apparent from the ensuing description and drawings in which:
FIG. 1 is an isometric view of a scanner for generating a pixel pattern representative of an image;
FIG. 2 is an example of the output from the scanner of FIG. 1 pixel values;
FIG. 3 is an example of a 1/2 tone screen function with threshold values;
FIG. 4 is an illustration showing the result of screening the scanner output of FIG. 2 with the screen of FIG. 3;
FIG. 5a is a schematic view of a predetermined quadrant template sequence;
FIG. 5b is an example of a quadrant nontemplate;
FIG. 6 illustrates exemplary established image values for the scanner output of FIG. 2 following screening using the quadrant bit sequence of FIG. 5a;
FIG. 7 illustrates an exemplary quadrant image value prediction;
FIG. 8 illustrates an exemplary established image value for the predicted quadrant of FIG. 7; and
FIG. 9 is a block diagram of the encoder/decoder apparatus of the present invention.
In the ensuing description, pixel refers to a picture element which may be defined as a scaler value of the gray scale of an image at a particular point of the image.
Referring to FIGS. 1-9 of the drawings, there is shown graphically the preferred steps that comprise the image data compressing system of the present invention. Initially, an original document 10 is scanned by a suitable line scanner 11 to produce pixel patterns 12 representative thereof. The value of the pixels as will be understood by those skilled in the art provides a gray scale representation of the image scanned. Exemplary pixel values are illustrated in FIG. 2, which for purposes of explanation may be considered as voltage levels.
The image pixel patterns 12 are screened electronically, the screen being illustrated herein by a template 16 of predetermined size and predetermined thresholds. A typical threshold pattern is shown in FIG. 3. Application of the threshold pattern of FIG. 3 to the pixel pattern 12 of FIG. 2 produces the binary pixel pattern 17 of FIG. 4. As will be understood, threshold values and patterns other than that illustrated may be contemplated.
To facilitate encoding of the image data and to reduce complexity, template 16 is in effect subdivided into discrete areas such as quadrants 18. The binary image pixel patterns 17 in quadrants 18 are compared with a predetermined quadrant template pixel patterns 19 18' such as shown in FIG. 5a, each pixel pattern 19 having a numerical value (herein termed established quadrant image values QV) assigned thereto. Matching of an actual binary image pixel pattern 17 with a template pixel pattern 19 produces the established quadrant image value QV therefor. See FIG. 6 for example.
A quadrant pixel pattern not falling in the pixel sequence 19 in FIG. 5a, as for example the quadrant pixel pattern 17' of FIG. 5b, is defined as a nontemplate pixel pattern 20. As can be understood, the use of quadrant templates and nontemplates pixel patterns 19, 20 reduces the number of potential image values, on a per quadrant basis, to 11, 10 quadrant template pixel pattern 19 and one quadrant nontemplate pixel pattern 20.
With the screened image pixel pattern 17 established for each quadrant, an image code is generated using quadrant prediction and comparison. Referring to FIG. 7, the image value of quadrant templates 18 are predicted on a per quadrant basis using the established image values of adjoining quadrants according to the following relationship:
QV=f (QV1, QV2) where
QV represents the predicted quadrant image value, and
QV represents the established quadrant image value.
In the example shown in FIG. 7, quadrant 30 (QV1) has a real image value of 5, and quadrant 31 (QV2) an image value of 3. The predicted image value (QV) for the adjoining quadrant 32 is 4.
Following quadrant image value prediction, the established image value (QV) for each quadrant is compared with the predicted image value (QV) of the quadrant in accordance with the following relationship:
e=QV -QV wherein
e represents the error
In the example shown, the established image value (QV) for quadrant 32 is 3 (see FIG. 8), the predicted image value (QV) is 4. The error (e) in this case would be -1.
In the case where the quadrant borders the image margin, such as quadrant 33 in FIG. 5, the areas outside the image margin are presumed to have a preset image value. In the example shown, the image value for the area outside the image margin is assumed to be zero.
The error signal so determined may be encoded using the code set out in Table I, the encoded image data together with unencoded nontemplate image data representing the original image 10 scanned in the form of a relatively highly compressed data stream. The image may thereafter be exactly reconstructed from the data by reversing the aforedescribed process.
TABLE I______________________________________ERROR VALUE CODE______________________________________ 0 0+1 10-1 110+2 11101-2 11110+3 111001-3 1111100+4 1111101-4 1111110+5 1111111-5 11100000+6 11100010-6 111000010+7 111000111-7 1110000110+8 1110000111-8 1110001101+9 11100011001-9 111000110000+10 1110001100010-10 1110001100011______________________________________
It is appreciated that other statistical encoders such as a run length encoder may be used instead to encode the aforedescribed error signal.
Referring to FIG. 9, image pixel data generated by scanner 11 is fed to screening device 50 wherein screen 16 subdivided into quadrant templates 18, is electronically superimposed onto the image pixel patterns 12. Screening device 50 is conventional as known to those skilled in the art. The screened data, in the form of a binary pixel pattern, is fed via data bus 52 to data storage unit 54. Storage unit 54 serves to progressively buffer the binary image data on a line by line basis to provide a three line quadrant output to row registers 62, 63, 64. Storage unit 54 incorporates series coupled line shift registers 58, 59, 60 operatively coupled to row registers 62, 63, 64 respectively.
The data output of registers 62, 63, 64 is fed to data reader and encoder 65. Encoder 65 comprises a random access memory (RAM) for mapping the actual image pixel patterns 17, 17' relative to the quadrant template patterns 19, 20. This establishes the actual quadrant image value (QV) of quadrant 18 and distinguishes template pixel patterns 19 from nontemplate pixel patterns 20 to provide, via control lead 67, a control signal to data output selector 98 as will appear.
The established quadrant image value (QV) from encoder 65 is fed via distributor register 70 to quadrant image predictor 74 and to scan line buffer storage 75. Buffer storage 75 holds the established quadrant image values pending use thereof as predictors for adjoining but succeeding quadrants, it being understood that encoder 65 operates on one quadrant line at a time. Predictor 74, which comprises a random access memory (RAM) predicts from the established image values of adjoining pairs of quadrants (i.e., quadrants 30, 31 in FIG. 6), a predicted quadrant image value (QV) for each quadrant 18. The predicted quadrant image values (QV) of predictor 74 are fed to holding register 84.
Established quadrant image values (QV) from encoder 74 are fed to holding register 82. Registers 82, 84 serve to synchronize the predicted and established quadrant image values (QV, QV) for comparison purposes.
The established and predicted quadrant image values (QV,QV) from registers 82, 84 are fed to arithmetic logic unit 88. Logic unit 88 compares the established quadrant image value (QV) with the predicted quadrant image value (QV) and generates an error signal representative of the degree of error (e) if any, therebetween. The signal from logic unit 88 is fed to holding register 94 and thereafter inputted to encoder 96. Encoder 96 comprises a random access memory (RAM) programmed to transform the error signal into a coded output in accordance with the parameters set forth in Table I. The coded signal from encoder 96 is transmitted via selector 98 to output bus 100.
As described, nontemplate pixel patterns 20, i.e., data whose bit sequence is different from the quadrant bit pattern programmed for, may occur. Image data in this category is transmitted via line serializing shift registers 102, 103, 104 directly to data output selector 98. The nontemplate quadrant data, on a control signal from encoder 65, is placed on output bus 100 in corrent line synchronization with the encoded error signals.
As will be understood, the compressed data may be used to reconstruct the original image by reversing the herein described process. More particularly, the encoded error signal may be decoded using Table I in reverse to provide an error signal, the value of which represents the magnitude of the error between the established quadrant image value (QV) and the predicted quadrant image value (QV). With the image values of the areas outside the image margin preset (i.e., zero), the predictive process may be reversed to determine the established quadrant image values. Using the established quadrant image values and the template/nontemplate quadrant pixel sequence of FIGS. 5a and 5b, the quadrant pixel pattern is exactly determined. This results in a binary representative of the original image.
While the invention has been described with reference to the structure disclosed, it is not confined to the details set forth, but is intended to cover such modifications or changes as may come within the scope of the following claims.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US3462547 *||Sep 6, 1966||Aug 19, 1969||Stanford Research Inst||Data processing system for signals obtained from a video scanner|
|US3573789 *||Dec 13, 1968||Apr 6, 1971||Ibm||Method and apparatus for increasing image resolution|
|US3769453 *||Aug 17, 1972||Oct 30, 1973||Ibm||Finite memory adaptive predictor|
|US3980809 *||Mar 6, 1975||Sep 14, 1976||International Business Machines Corporation||Encoding logic for reduced bandwidth pictorial data transmission system|
|US4013828 *||Feb 20, 1976||Mar 22, 1977||Bell Telephone Laboratories, Incorporated||Method and arrangement for reducing the bandwidth and/or time required to transmit a dithered image|
|US4028731 *||Sep 29, 1975||Jun 7, 1977||International Business Machines Corporation||Apparatus for compression coding using cross-array correlation between two-dimensional matrices derived from two-valued digital images|
|US4047152 *||Nov 3, 1976||Sep 6, 1977||Elettronica San Giorgio - Elsag - S.P.A.||Electronic character-reading system|
|US4144547 *||Apr 4, 1977||Mar 13, 1979||Xerox Corporation||Apparatus and method for encoding halftone and line copy data|
|US4150400 *||Mar 31, 1977||Apr 17, 1979||International Business Machines Corporation||Methods of a coarse-scan/fine-print character reproduction with compression|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US4245248 *||Apr 4, 1979||Jan 13, 1981||Bell Telephone Laboratories, Incorporated||Motion estimation and encoding of video signals in the transform domain|
|US4313194 *||Dec 27, 1979||Jan 26, 1982||Maitre Xavier C||Transmission system of frequency division multiplex signals on digital links|
|US4417276 *||Apr 16, 1981||Nov 22, 1983||Medtronic, Inc.||Video to digital converter|
|US4475127 *||Feb 23, 1982||Oct 2, 1984||Nippon Electric Co., Ltd.||System for transmitting a video signal with short runs avoided in a signal encoded from the video signal|
|US4506342 *||Nov 2, 1981||Mar 19, 1985||Tokyo Shibaura Denki Kabushiki Kaisha||Document information filing system|
|US4603431 *||Mar 14, 1983||Jul 29, 1986||Ana Tech Corporation||Method and apparatus for vectorizing documents and symbol recognition|
|US4633325 *||Sep 4, 1984||Dec 30, 1986||Nec Corporation||Adaptive predictive encoding and/or decoding apparatus|
|US4654484 *||Jul 21, 1983||Mar 31, 1987||Interand Corporation||Video compression/expansion system|
|US4706077 *||Sep 8, 1981||Nov 10, 1987||Xerox Corporation||Halftoning implementation for interactive image editing|
|US4720743 *||Nov 12, 1985||Jan 19, 1988||NEC Corporation and Nippon Telegraph and Telephone Corporation||Predictine coding/decoding system for block-formed picture signals|
|US4750044 *||Feb 5, 1987||Jun 7, 1988||Minolta Camera Kabushiki Kaisha||Image editor|
|US4758886 *||Jul 24, 1986||Jul 19, 1988||Minnesota Mining And Manufacturing Company||Optimal color half-tone patterns for raster-scan images|
|US4916545 *||Nov 16, 1988||Apr 10, 1990||Eastman Kodak Company||Electronic graphic arts screener that suppresses Moire patterns using pseudo-random font selection|
|US4918622 *||Nov 16, 1988||Apr 17, 1990||Eastman Kodak Company||Electronic graphic arts screener|
|US4977458 *||Nov 16, 1988||Dec 11, 1990||Eastman Kodak Company||Apparatus for addressing a font to suppress Moire patterns occurring thereby and a method for use therein|
|US5020120 *||Dec 22, 1989||May 28, 1991||Eastman Kodak Company||Methods for reducing quantization error in hierarchical decomposition and reconstruction schemes|
|US5181126 *||Dec 9, 1991||Jan 19, 1993||Murata Kikai Kabushiki Kaisha||Reading device for facsimile apparatus|
|US5274473 *||Dec 16, 1991||Dec 28, 1993||Intergraph Corporation||Rapid variable angle digital screening|
|US5489991 *||Oct 9, 1992||Feb 6, 1996||International Business Machines Corporation||Method for generating digital halftone images|
|US5508733 *||Feb 12, 1993||Apr 16, 1996||Kassatly; L. Samuel A.||Method and apparatus for selectively receiving and storing a plurality of video signals|
|US5691777 *||Jan 27, 1992||Nov 25, 1997||Kassatly; Lord Samuel Anthony||Method and apparatus for simultaneous compression of video, audio and data signals|
|US5767913 *||May 25, 1995||Jun 16, 1998||Kassatly; Lord Samuel Anthony||Mapping system for producing event identifying codes|
|US5768517 *||May 25, 1995||Jun 16, 1998||Kassatly; Samuel Anthony||Paperless publication distribution and retrieval system|
|US5790177 *||May 25, 1995||Aug 4, 1998||Kassatly; Samuel Anthony||Digital signal recording/reproduction apparatus and method|
|US6201614||Dec 24, 1998||Mar 13, 2001||Electronics For Imaging, Inc.||Dithered image compression system|
|US6222641||Jul 1, 1998||Apr 24, 2001||Electronics For Imaging, Inc.||Method and apparatus for image descreening|
|US6330362||Nov 12, 1996||Dec 11, 2001||Texas Instruments Incorporated||Compression for multi-level screened images|
|US6450611||Feb 14, 2001||Sep 17, 2002||Lexmark International, Inc.||Ink jet swath prediction method|
|US6757436||Dec 19, 2002||Jun 29, 2004||Electroncs For Imaging, Inc.||Methods and apparatus for data compression based on modeling schemes|
|US6757440||Jul 18, 2002||Jun 29, 2004||Electronics For Imaging, Inc.||Methods and apparatus for data compression|
|US6798543||Sep 13, 2000||Sep 28, 2004||Sharp Laboratories Of America Inc.||Method and apparatus for compression and decompression of halftoned images|
|US7058231||Jun 1, 2004||Jun 6, 2006||Electronics For Imaging, Inc.||Methods and apparatus for data compression with a hybrid context|
|US7460721||May 4, 2006||Dec 2, 2008||Electronics For Imaging, Inc.||Methods and apparatus for data compression with a hybrid context|
|US20040223654 *||Jun 1, 2004||Nov 11, 2004||Peters Michael Alan||Methods and apparatus for data compression|
|EP0051259A2 *||Oct 27, 1981||May 12, 1982||Kabushiki Kaisha Toshiba||Document information filing system|
|EP0066697A1 *||Apr 21, 1982||Dec 15, 1982||International Business Machines Corporation||A method and system for compressing grey scale image data|
|EP0690613A2 *||Jun 27, 1995||Jan 3, 1996||Seiko Epson Corporation||System for compressing a half-tone image and method therefor|
|WO1998058452A2 *||Jun 4, 1998||Dec 23, 1998||Electronics For Imaging, Inc.||Method and apparatus for data compression|
|WO1998058452A3 *||Jun 4, 1998||Mar 11, 1999||Electronics For Imaging Inc||Method and apparatus for data compression|
|WO2000040003A1 *||Oct 11, 1999||Jul 6, 2000||Electronics For Imaging, Inc.||Dithered image compression system|
|U.S. Classification||382/238, 358/426.12, 375/240.12, 375/240.24|